Birds learn to sing by using auditory feedback to match their own vocal output against their memory of a song model. The motor system for the control of song is a discrete neural pathway, and an auditory input to the system has recently been identified. This program of research seeks to relate (1) The nature of the auditory input to the song system, and its relevance for vocal learning, to (2) structural manifestations of synaptic plasticity during vocal learning. The avian song system should be a preparation in which it will be possible to ask whether or not groups of neurons come to encode learned behaviors through modification of their connectivity. The proposed studies will make use of intracellular recording and staining of single neurons by injection of horseradish peroxidase.